Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    text processing/tokenizing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     infirm
    -0.08
    �
    -0.08
     آدم
    -0.08
     عندما
    -0.08
     terminate
    -0.07
     deductions
    -0.07
     wager
    -0.07
     hired
    -0.07
     Governance
    -0.07
     rendel
    -0.07
    POSITIVE LOGITS
     strings
    0.10
     Strings
    0.10
    _strings
    0.09
    Strings
    0.08
    _TEXT
    0.08
     речи
    0.08
    ”↵↵
    0.08
    _ENCOD
    0.08
    STRING
    0.08
    strings
    0.08
    Act Density 0.004%

    No Known Activations